Intelligent networked vehicle safety protection system and method based on text watermarking

Abstract

This application provides a text watermark-based security protection system and method for intelligent connected vehicles, relating to the field of information hiding technology. The system includes a watermark generation module, a watermark embedding module, an in-vehicle data interaction module, a watermark extraction module, a security verification module, and a cloud management module. The watermark generation module generates a unique text watermark through hash operations; the watermark embedding module embeds the watermark into text data, outputting watermarked text data; the in-vehicle data interaction module transmits the watermarked text data; the watermark extraction module extracts the watermark information from the watermarked text data; the security verification module verifies the data; and the cloud management module stores identity information, watermark templates, and verification rules, and synchronizes the watermark data with the verification results. This application can address the current security risks faced by intelligent connected vehicles, such as network attacks, data tampering, and command forgery, improving the security and reliability of intelligent connected vehicle operation.

Description

Technical Field

[0001] This application relates to the field of information hiding technology, and in particular to a security protection system and method for intelligent connected vehicles based on text watermarking. Background Technology

[0002] With the rapid development of automotive intelligence and connectivity technologies, intelligent connected vehicles have become a core direction for the transformation and upgrading of the automotive industry. Intelligent connected vehicles integrate onboard sensors, communication modules, computing platforms, and autonomous driving algorithms, enabling multi-dimensional information interaction between vehicles (V2V), between vehicles and roads (V2I), and between vehicles and the cloud (V2C), significantly improving the driving experience and traffic efficiency. However, the security risks of intelligent connected vehicles are also becoming increasingly prominent, and traditional "plug-and-play" security defense strategies are insufficient to effectively cope with increasingly complex cyber threats.

[0003] Specifically, intelligent connected vehicles suffer from the following technical deficiencies in security protection: First, control commands such as acceleration, braking, and steering, as well as status data such as vehicle speed, tire pressure, and fault information, are easily intercepted, tampered with, or forged by attackers when transmitted via wireless communication links, potentially leading to vehicle malfunctions and safety accidents. Second, text-based data generated on-board, such as fault logs, driving records, and OTA upgrade instructions, lacks effective identification and traceability mechanisms. Once data leakage, tampering, or injection of false data occurs, it is difficult to trace the data source and traces of tampering. Third, current node authentication in intelligent connected vehicles largely relies on cryptographic encryption technology. These encryption algorithms are vulnerable to cracking, and the lack of identity verification methods tied to the data itself makes it difficult to prevent malicious attacks from spoofed nodes. Fourth, existing security technologies often employ complex encryption algorithms, increasing the computational burden on the on-board computing platform, affecting the real-time performance of command transmission and data processing, and making it difficult to deeply adapt to the transmission and storage scenarios of on-board text data.

[0004] Text watermarking technology is an important branch of information hiding technology. It can embed specific identifying information, i.e., a watermark, into text data without affecting the normal readability and use of the text. The watermark can be extracted when needed, thereby enabling data authentication, traceability, and integrity verification. Currently, text watermarking technology has been applied in areas such as document copyright protection and data traceability. However, there is no technical solution that deeply integrates it with the security of intelligent connected vehicles, thus failing to address the unique command security, data traceability, and authentication requirements of intelligent connected vehicles.

[0005] Furthermore, existing text watermarking technologies are mostly designed for general text and lack adaptability to text data in intelligent connected vehicles, such as short command text and structured log text. The embedded watermarks lack robustness and are susceptible to failure due to factors such as in-vehicle communication noise, data compression, and format conversion. Simultaneously, the watermark embedding and extraction processes do not consider the real-time requirements of intelligent connected vehicles, making it difficult to meet the rapid processing demands of in-vehicle scenarios. Therefore, there is an urgent need for a system and method that can deeply integrate text watermarking technology with the security requirements of intelligent connected vehicles to solve the aforementioned technical problems. Summary of the Invention

[0006] This application provides a text watermark-based intelligent connected vehicle security protection system and method to solve the problems of insufficient security protection for intelligent connected vehicles and poor adaptability of text watermark technology in the prior art. By deeply integrating text watermark with the instruction transmission, data storage and identity authentication of intelligent connected vehicles, it realizes instruction tamper-proof, data traceability and identity authentication, while ensuring the real-time and robustness of watermark embedding and extraction, thereby improving the overall security level of intelligent connected vehicles without affecting the normal operation of the vehicle system.

[0007] To achieve the above objectives, this application adopts the following technical solution: In the first aspect, this application provides a smart connected vehicle security protection system based on text watermarking, including a watermark generation module, a watermark embedding module, an in-vehicle data interaction module, a watermark extraction module, a security verification module, and a cloud management module; The watermark generation module is communicatively connected to the cloud management module and is used to generate a unique corresponding text watermark through hash operation based on the identity information, instruction type and security requirements of the intelligent connected vehicle. The watermark embedding module is connected to the watermark generation module and the vehicle data interaction module respectively, and is used to embed the generated text watermark into the text data of the intelligent connected vehicle and output watermarked text data. The vehicle data interaction module is connected to the watermark embedding module, the watermark extraction module, the vehicle sensor, and the vehicle control unit, and is used to transmit the watermarked text data and receive external feedback data. The watermark extraction module is connected to the vehicle data interaction module and the security verification module, and is used to extract watermark information from the received watermarked text data. The security verification module is connected to the watermark extraction module, the cloud management module and the vehicle control unit, and is used to verify the extracted watermark information to determine whether the data has been tampered with, whether the vehicle identification is consistent with the vehicle identification stored in the cloud, and whether the timestamp is within a preset valid range. The cloud management module is connected to the watermark generation module and the security verification module, and is used to store the identity information, watermark template, and verification rules of the intelligent connected vehicle, and to synchronize watermark data and verification results.

[0008] As a preferred technical solution, the watermark generation module generates the text watermark by combining encryption algorithms and hash operations. The text watermark includes a unique vehicle identifier, a data type identifier, a timestamp, and a verification code. The hash operation is represented as: (1) in, This is the watermark hash value. The unique identifier for intelligent connected vehicles. For timestamps, For data type identification, For an improved MD5 hash function, A unique key assigned to the cloud management module. The length of the watermark. This is an XOR operation.

[0009] As a preferred technical solution, the watermark embedding module adopts a differentiated embedding strategy for different types of text data. For short instruction text, the text watermark is embedded in punctuation marks or spaces; for structured log text, the text watermark is embedded in redundant bits of log fields. The embedding process of the watermark embedding module is represented as follows: (2) in, For watermarked text data, The original text data, This is the original watermark information. For embedding strength, This is the watermark embedding function.

[0010] As a preferred technical solution, the watermark extraction module adopts an extraction algorithm corresponding to the embedding algorithm, locates the watermark embedding position according to the type of text data, and extracts the watermark information; The watermark extraction process of the module is described as follows: (3) in, For the extracted watermark information, This is the watermark extraction function. For watermarked text data, For embedding strength.

[0011] As a preferred technical solution, the security verification module will extract the watermark information. The original watermark information stored in the cloud management module The comparison is performed to verify whether the vehicle identifier in the watermark is consistent with the vehicle identifier stored in the cloud and whether the timestamp is within the preset valid range. like If the vehicle identifier matches the vehicle identifier stored in the cloud and the timestamp is within a preset valid range, the verification passes; otherwise, the verification fails and a security warning is triggered. This is the preset allowable error.

[0012] Secondly, this application provides a method for security protection of intelligent connected vehicles based on text watermarking, the method comprising: A unique key is assigned to the intelligent connected vehicle through the cloud, the vehicle's unique identifier, current timestamp, and text data type identifier to be processed are obtained, a unique text watermark is generated through hash operation, and the text watermark is stored in the cloud; The vehicle-mounted terminal acquires the original text data, dynamically adjusts the embedding strength according to the text data type, and embeds the text watermark into the original text data to generate watermarked text data. The watermarked text data is transmitted to the target node, and the data is compressed and encrypted during the transmission process; The target node receives the watermarked text data and extracts the watermark information according to the text data type. The original watermark information is obtained from the cloud. The extracted watermark information is compared with the original watermark information. At the same time, it is verified whether the vehicle identification is consistent with the vehicle identification stored in the cloud and whether the timestamp is within the preset valid range. The verification result is then output. If the verification passes, the target node executes relevant instructions or stores data; if the verification fails, a security warning is triggered, abnormal data transmission is cut off, execution of relevant instructions is prohibited, and abnormal information is uploaded to the cloud.

[0013] As a preferred technical solution, the hash operation is represented as follows: (1) in, This is the watermark hash value. The unique identifier for intelligent connected vehicles. For timestamps, For data type identification, For an improved MD5 hash function, A unique key assigned to the cloud management module. The length of the watermark. This is an XOR operation.

[0014] As a preferred technical solution, the watermark embedding process is represented as follows: (2) in, For watermarked text data, The original text data, This is the original watermark information. For embedding strength, This is the watermark embedding function.

[0015] As a preferred technical solution, the watermark extraction process is represented as follows: (3) in, For the extracted watermark information, This is the watermark extraction function. For watermarked text data, For embedding strength.

[0016] As a preferred technical solution, the security verification rule is: if the extracted watermark information With the original watermark information satisfy If the vehicle identifier matches the vehicle identifier stored in the cloud and the timestamp is within a preset valid range, the verification passes; otherwise, the verification fails. This is the preset allowable error.

[0017] The intelligent connected vehicle security protection system and method based on text watermarking provided in this application have at least the following beneficial effects: First, comprehensive security protection. This application deeply integrates text watermarking with the command transmission, data storage, and identity authentication of intelligent connected vehicles. A watermark generation module uses hash calculations to generate a unique text watermark containing a unique vehicle identifier, timestamp, and verification code. A security verification module performs integrity comparisons and verifies the validity of the vehicle identifier and timestamp, achieving tamper-proof commands, traceable data, and verifiable identity. This solution effectively resists network attacks such as interception, tampering, and forgery, addressing the core security vulnerabilities faced by intelligent connected vehicles, while also complying with the mandatory provisions of national standards such as the technical requirements for information security of automotive systems for intelligent connected vehicles.

[0018] Second, it boasts strong adaptability. This application employs differentiated watermark embedding and extraction strategies for different types of text data from intelligent connected vehicles, including short command text and structured log text. The watermark embedding module dynamically adjusts the embedding strength based on the text type, with a value ranging from 0.1 to 0.3. Watermarks are embedded in non-semantic key positions in short command text, such as punctuation marks or spaces, or in redundant positions in log fields. This embedding process does not affect the normal readability and usability of the text, and the watermark extraction process is robust against interference from in-vehicle communication noise, data compression, and format conversion, fully adapting to the specific needs of in-vehicle scenarios.

[0019] Third, it offers excellent real-time performance. The watermark embedding and extraction process in this application employs a lightweight algorithm with low computational complexity for the embedding and extraction functions. Combined with the 5G-V2X communication protocol used in the vehicle data interaction module, the watermarked text data is compressed and encrypted to ensure that the data transmission and processing latency is no more than 100ms, meeting the real-time requirements of intelligent connected vehicles without significantly increasing the computing power burden on the vehicle computing platform.

[0020] Fourth, it boasts strong scalability. This application's system adopts a modular design, with the watermark generation module, watermark embedding module, in-vehicle data interaction module, watermark extraction module, security verification module, and cloud management module operating independently and being flexibly expandable. This allows it to adapt to the safety requirements of different levels of intelligent connected vehicles, such as L2 or L3 level assisted driving vehicles. Furthermore, the system can seamlessly integrate with existing in-vehicle safety systems and cloud management platforms, facilitating industrialization and widespread application.

[0021] Fifth, precise source tracing. The text watermark generated by this application contains key information such as the vehicle's unique identifier (VIN code), timestamp, and data type identifier. In the event of data anomalies, the source of the data and traces of tampering can be quickly traced by extracting the above information from the watermark, providing strong evidence for safety accident investigation and liability determination, and effectively solving the current problem of ambiguous division of responsibilities in intelligent connected vehicle accidents. Attached Figure Description

[0022] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0023] Figure 1 A flowchart illustrating a design method for a landing gear torsion arm provided in this application embodiment; Figure 2 A flowchart illustrating the watermark embedding and extraction process provided in this application embodiment; Figure 3 The flowchart of a text watermark-based intelligent connected vehicle security protection method provided in this application embodiment Figure 1 ; Figure 4 The flowchart of a text watermark-based intelligent connected vehicle security protection method provided in this application embodiment Figure 2 .

[0024] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0025] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0026] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0027] Example 1: This application provides a text watermark-based intelligent connected vehicle security protection system, such as... Figure 1As shown, the system includes a watermark generation module 1, a watermark embedding module 2, an in-vehicle data interaction module 3, a watermark extraction module 4, a security verification module 5, and a cloud management module 6. The watermark generation module 1 is communicatively connected to the cloud management module 6 and is used to generate a unique text watermark based on the intelligent connected vehicle's identity information, command type, and security requirements through hash calculations. The watermark embedding module 2 is connected to both the watermark generation module 1 and the in-vehicle data interaction module 3, and is used to embed the generated text watermark into the intelligent connected vehicle's text data, outputting watermarked text data. The in-vehicle data interaction module 3 is connected to the watermark embedding module 2, the watermark extraction module 4, and the in-vehicle sensors and control units. The system is used to transmit watermarked text data and receive external feedback data. The watermark extraction module 4 is connected to the vehicle data interaction module 3 and the security verification module 5 to extract watermark information from the received watermarked text data. The security verification module 5 is connected to the watermark extraction module 4, the cloud management module 6 and the vehicle control unit to verify the extracted watermark information, determine whether the data has been tampered with, whether the vehicle identification is consistent with the vehicle identification stored in the cloud, and whether the timestamp is within the preset valid range. The cloud management module 6 is connected to the watermark generation module 1 and the security verification module 5 to store the identity information, watermark template and verification rules of the intelligent connected vehicle, and synchronize the watermark data and verification results.

[0028] Specifically, such as Figure 1 As shown, the overall system architecture is divided into an external entity layer and an intelligent connected vehicle local layer. The external entity layer includes a cloud management module 6 and a roadside node 9. The cloud management module 6 and the roadside node 9 achieve unified configuration of global security parameters through identity information and key / rule synchronization. The roadside node 9 and the intelligent connected vehicle local layer achieve bidirectional transmission of feedback traffic data and watermarked data through 5G-V2X communication.

[0029] Within the local layer of the intelligent connected vehicle, the cloud management module 6 first sends the identity information, the complete watermark template, the embedding strength, and the unique key to the watermark generation module 1. Simultaneously, the watermark generation module 1 obtains the vehicle's unique identifier (VIN) and the current timestamp. T Data type identifier of the text to be processed D and the text to be processed S A unique text watermark is generated through hash operations. The hash operation uses an improved MD5 hash function, which performs an XOR operation on the vehicle's unique identifier, timestamp, and the unique key allocated in the cloud, and then takes the modulo to obtain the watermark hash value. The generated text watermark... The data is transmitted to the watermark embedding module 2, which simultaneously receives the original text S transmitted by the vehicle data interaction module 3 and the embedding strength from the cloud management module 6. αAn embedding algorithm based on text semantic invariance is used to embed the watermark into the original text, generating watermarked text. W Embedding strength α The value ranges from 0.1 to 0.3, dynamically adjusted according to the length of the original text. For short instruction text, the watermark is embedded in non-semantic key positions such as punctuation marks and spaces; for structured log text, the watermark is embedded in redundant bits of log fields, ensuring that the readability and integrity of the embedded text are not affected. The generated watermarked text... W The data is transmitted to the vehicle-mounted data interaction module 3. As the hub for local data interaction, the vehicle-mounted data interaction module 3 receives feedback data and safety commands collected by the vehicle-mounted sensors 7, and receives feedback traffic data and watermarked data transmitted by the roadside node 9 via 5G-V2X communication. Simultaneously, it transmits the received raw text data to the watermark embedding module 2 for watermarking processing, generating watermarked text. W The data is transmitted to the vehicle control unit 8 and the watermark extraction module 4, respectively.

[0030] The vehicle-mounted data interaction module 3 uses the 5G-V2X communication protocol to transmit all data. During transmission, the data is compressed and encrypted to ensure a transmission latency of no more than 100ms, meeting the real-time requirements of intelligent connected vehicles. The watermark extraction module 4 receives the watermarked text W transmitted by the vehicle-mounted data interaction module 3, uses a reverse extraction algorithm corresponding to the embedding algorithm to locate the watermark embedding position, and extracts the watermark information. The extraction process satisfies The inverse function of E acts on W and α The inverse function of E is the watermark extraction function. The embedding strength used in the extraction process is consistent with that used in the embedding process. This extraction algorithm can resist external interference such as vehicle communication noise, data compression, and format conversion, ensuring the accuracy of watermark extraction. The extracted watermark information... The data is transmitted to the security verification module 5, which simultaneously retrieves the verification rules and the original watermark from the cloud management module 6. The extracted watermark information With the original watermark A comparison verification is performed. If the two match perfectly and the vehicle identifier and timestamp in the watermark are valid, the verification is considered successful. The security verification module 5 sends a decision signal to the vehicle control unit 8, allowing the vehicle control unit 8 to execute control commands, thereby driving the wheel / power system and wheel / braking system to complete the corresponding actions. If the comparison verification fails, the verification is considered unsuccessful. The security verification module 5 immediately triggers an alarm and uploads an anomaly report to the cloud management module 6, while simultaneously prohibiting the sending of execution commands to the vehicle control unit 8. After receiving the anomaly report, the cloud management module 6 records key information such as the anomaly type, occurrence time, and vehicle identifier, generates a security report, and issues security instructions to the vehicle nodes to guide emergency handling. At the same time, it continuously synchronizes identity information, keys, and verification rules with the roadside nodes 9 to achieve global security control of the entire intelligent connected vehicle system.

[0031] In some embodiments, the watermark generation module 1 uses a combination of encryption algorithm and hash operation to generate a unique and unforgeable text watermark. The watermark information includes a unique vehicle identifier (VIN code), a data type identifier, a timestamp, and a verification code. The watermark generation process introduces mathematical formula (1) for hash operation to ensure the uniqueness of the watermark. (1) in, This is the watermark hash value. VIN This is the unique identifier for intelligent connected vehicles. T For timestamps, D For data type identification, For an improved MD5 hash function, K A unique key assigned to the cloud. n This is the watermark length (values ​​are 16 or 32, adapted to the length of in-vehicle text data). This is an XOR operation.

[0032] It should be noted that for the improved MD5 hash function Its expression and implementation are as follows: The function input is M = VIN ⊕ T ⊕ K ,in VIN It serves as a unique identifier for the vehicle. T For timestamps, K A unique key assigned to the cloud. (Function) H m (∙) Through Padding( M ) Analyze the input message according to the MD5 standard format M Padding is performed to ensure its length is a multiple of 512 bits; the compression function of the round key is introduced sequentially through four rounds. F 1~F 4. Perform iterative calculations, where F 1~ F 4 are respectively: F 1( X , Y , Z )=( X ∧ Y )∨(! X ∧ Z )⊕ K 1; F 2( X , Y , Z )=( X ∧ Z )∨( Y ∧! Z )⊕ K 2; F 3( X , Y , Z )= X ⊕ Y ⊕ Z ⊕ K 3; F 4( X , Y , Z )= Y ⊕( X ∨! Z )⊕ K 4; In this context, the symbols ! represent the logical NOT operation, which inverts the operand; ∧ represents the logical AND operation; ∨ represents the logical OR operation; and ⊕ represents the bitwise XOR operation. X , Y , Z This is a 32-bit state variable used in the MD5 compression process; K 1~ K 4 is the round key generated based on the cloud key.

[0033] Take the low-order bits of the 128-bit digest of the MD5 output. n Bit, n Taking 16 or 32, we get H m ( M ), which is the watermark hash value adapted to the length of the in-vehicle text data.

[0034] In some embodiments, the watermark embedding module 2 employs an embedding algorithm based on text semantic invariance, and adopts differentiated embedding strategies for different types of text data (short instruction text, structured log text) of intelligent connected vehicles: for short instruction text (such as "accelerate" and "brake"), the watermark is embedded in non-semantic key positions such as punctuation marks and spaces in the text; for structured log text, the watermark is embedded in redundant positions of the log fields. The embedding process satisfies formula (2) to ensure the readability and integrity of the embedded text: (2) in, For watermarked text data, S The original text data, For the generated watermark information, Embedding strength (value range: 0.1~0.3, dynamically adjusted according to text length), This is the watermark embedding function.

[0035] In some embodiments, the vehicle data interaction module 3 adopts the 5G-V2X communication protocol to realize the real-time transmission of watermarked text data between vehicle nodes, roadside nodes, and cloud nodes. During the transmission process, the data is compressed to ensure that the transmission delay is ≤100ms, which meets the real-time requirements of intelligent connected vehicles. At the same time, it receives status data collected by vehicle sensors and traffic data fed back by roadside nodes, and transmits them to the watermark embedding module for watermark embedding.

[0036] In some embodiments, the watermark extraction module 4 employs an extraction algorithm corresponding to the embedding algorithm, locates the watermark embedding position according to the text data type, and extracts the watermark information. The extraction process satisfies formula (3), which can resist interference from communication noise, data compression, format conversion, etc., and ensure the accuracy of watermark extraction. (3) in, For the extracted watermark information, This is the watermark extraction function. For the received watermarked text data, The embedding strength (consistent with the embedding process).

[0037] In some embodiments, the security verification module 5 will extract the watermark information. Original watermark information stored in the cloud management module A comparison is performed, and the legality of the vehicle identification and timestamp in the watermark is verified. The verification rules are as follows: like ( If the error margin is 0 (meaning a perfect match) and the vehicle identifier and timestamp are both valid, the verification passes, indicating that the data has not been tampered with and the identity is legitimate, allowing the execution of relevant instructions or the storage of data. If the verification fails, the data is determined to have been tampered with or the identity is illegitimate, immediately triggering a security warning, cutting off abnormal data transmission, prohibiting the execution of relevant instructions, and uploading the abnormal information to the cloud management module.

[0038] In some embodiments, the cloud management module 6 adopts a distributed storage architecture to store the identity information, watermark templates, keys and verification rules of all intelligent connected vehicles, and synchronizes the watermark data and verification results of the vehicle nodes in real time. When a security anomaly occurs, it records the anomaly information (anomaly type, time, vehicle identifier), generates a security report, and sends security instructions to the vehicle nodes to guide the vehicle system to perform emergency handling.

[0039] As an exemplary implementation, please refer to Figure 2 The specific workflow of each module in this text watermark-based intelligent connected vehicle security protection system is as follows.

[0040] Watermark generation module 1: Using an improved MD5 hash function, obtain the VIN code of the intelligent connected vehicle (e.g., "LSVCC2A40NN000001"), the current timestamp T (e.g., "20260306105800"), and the data type identifier D (control command identifier is "01", fault log identifier is "02"). The unique key K allocated by the cloud is "1234567890ABCDEF". Calculate the watermark hash value using formula (1):

[0041] The calculated watermark hash value is "5A7B9C3D". Combined with the vehicle VIN code and timestamp, a complete watermark is generated. The value is “LSVCC2A40NN000001-20260306105800-5A7B9C3D”, and it is uploaded to the cloud management module 6 for storage.

[0042] Watermark embedding module 2 adjusts the embedding strength for the control command text S of intelligent connected vehicles (such as "accelerate to 60km / h"). =0.2, the watermark is removed using formula (2). The embedded text contains spaces and punctuation marks, resulting in watermarked text. The embedding process for "accelerating to 60km / h (LSVCC2A40NN000001-20260306105800-5A7B9C3D)" does not affect the normal reading and execution of the instruction.

[0043] The vehicle-mounted data interaction module 3 uses the 5G-V2X communication protocol to transmit watermarked text. The data is transmitted to the vehicle control unit 8. During the transmission process, the LZ77 compression algorithm is used to compress the data, and the transmission delay is 80ms, which meets the real-time requirements. At the same time, the vehicle speed data collected by the vehicle sensor 7 is received and transmitted to the watermark embedding module 2 for watermark embedding.

[0044] Watermark extraction module 4 receives watermarked text. Watermark information is extracted using formula (3). The value "LSVCC2A40NN000001-20260306105800-5A7B9C3D" is transmitted to security verification module 5.

[0045] The security verification module 5 obtains the vehicle's original watermark from the cloud management module 6. ,Will and Upon comparison, the two are completely identical. Furthermore, the vehicle's VIN code and timestamp are both valid, so the verification is deemed successful, and the vehicle control unit 8 is allowed to execute the command to "accelerate to 60km / h".

[0046] Cloud management module 6 stores the vehicle's VIN code and watermark. Key K It synchronizes verification results in real time, records instruction execution, and achieves global security control.

[0047] Example 2: This invention provides a text-watermark-based intelligent connected vehicle security protection method, which can be implemented based on the text-watermark-based intelligent connected vehicle security protection system described in any of the embodiments of Embodiment 1, such as... Figure 3 As shown, the method includes the following steps S10-S60.

[0048] S10: Assign a unique key to the intelligent connected vehicle through the cloud, obtain the vehicle's unique identifier, current timestamp, and text data type identifier to be processed, generate a unique text watermark through hash operation, and store the text watermark in the cloud.

[0049] For example, the cloud management module 6 assigns a unique key K = "ABCDEF1234567890" to the intelligent connected vehicle, and the watermark generation module 1 obtains the vehicle VIN code "LSVCC2A40NN000002", timestamp T = "20260306110000", and data type identifier D = "02" (fault log), and calculates the watermark hash value using formula (1) to generate the watermark. =“LSVCC2A40NN000002-20260306110000-8F4E2D6C”, uploaded to cloud management module 6 storage.

[0050] S20: The vehicle-mounted terminal acquires the original text data, dynamically adjusts the embedding strength according to the text data type, embeds the text watermark into the original text data, and generates watermarked text data.

[0051] For example, the vehicle data interaction module 3 obtains the original text of the fault log S = "20260306105900, abnormal tire pressure, value 2.0 bar", transmits it to the watermark embedding module 2, adjusts the embedding strength α = 0.3, and embeds the watermark using formula (2). Embedded redundant fields in logs to generate watermarked text =“20260306105900, abnormal tire pressure, value 2.0 bar, watermark: LSVCC2A40NN000002-20260306110000-8F4E2D6C”.

[0052] S30: Transmit the watermarked text data to the target node, and compress and encrypt the data during the transmission process.

[0053] For example, the vehicle-mounted data interaction module 3 uses the 5G-V2X communication protocol to transmit watermarked text. The data is transmitted to the cloud management module 6 with a transmission latency of 95ms. During the transmission process, the data is encrypted using the AES-128 encryption algorithm.

[0054] S40: The target node receives watermarked text data and extracts the watermark information based on the text data type.

[0055] For example, the watermark extraction module 4 of the cloud management module 6 receives watermarked text. Watermarks can be extracted using formula (3). =“LSVCC2A40NN000002-20260306110000-8F4E2D6C”, transmitted to security verification module 5.

[0056] S50: Obtain the corresponding original watermark information from the cloud, compare the extracted watermark information with the original watermark information, and verify whether the vehicle identification is consistent with the vehicle identification stored in the cloud and whether the timestamp is within the preset valid range, and output the verification result.

[0057] For example, the security verification module 5 will With cloud storage The two are completely identical, and the vehicle VIN code and timestamp are valid. The verification is successful, and the cloud management module 6 stores the fault log.

[0058] S60: If the verification passes, the target node executes relevant instructions or stores data; if the verification fails, a security warning is triggered, abnormal data transmission is cut off, execution of relevant instructions is prohibited, and abnormal information is uploaded to the cloud.

[0059] For example, if an attacker tampers with the watermarked text The watermark extracted by watermark extraction module 4 is for the following information: "20260306105900, tire pressure normal, value 2.5 bar, watermark: LSVCC2A40NN000002-20260306110000-8F4E2D6C". With the original The data is consistent, but the text content does not match the data type and timestamp in the watermark. The security verification module 5 determines that the verification fails and immediately triggers a security warning. The abnormal information (tampering type, time, vehicle identification) is uploaded to the cloud management module 6. The cloud sends a tire pressure re-check instruction to the vehicle node and records the abnormal log at the same time.

[0060] In another implementation, such as Figure 4 As shown, the intelligent connected vehicle security protection method based on text watermarking can be implemented through the following steps S1-S6.

[0061] S1: Watermark generation.

[0062] Obtain the vehicle identifier (VIN), timestamp (T), and data type (D); generate a unique hash watermark using the improved MD5 formula (1). Upload Store in the cloud.

[0063] S2: Watermark embedding.

[0064] In-vehicle interaction module obtains raw text S Dynamically adjust intensity α Using semantically invariant embedding formula (2) to W Embedded S Generate watermarked text .

[0065] S3: Data transmission.

[0066] Using the 5G-V2X protocol, After compression and encryption, the data is transmitted to the target nodes (vehicle control unit, roadside node, cloud). S4: Watermark extraction.

[0067] Target node receives Locate the watermark position and extract the watermark using the inverse formula (3). .

[0068] S5: Security check.

[0069] Security inspection module 5 retrieves the original data from cloud management module 6. ;contrast and Verify the legality of vehicle identification and timestamp.

[0070] S6: Determine if the condition is met. : If so, proceed to S6-1: Execution and Storage, and execute the target node's instructions (such as acceleration and braking); store the structured log text in the cloud; If not, proceed to S6-2: Emergency response triggers security alert; abnormal data transmission is cut off; abnormal information is uploaded to the cloud.

[0071] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A smart connected vehicle security protection system based on text watermarking, characterized in that, It includes a watermark generation module, a watermark embedding module, an in-vehicle data interaction module, a watermark extraction module, a security verification module, and a cloud management module; The watermark generation module is communicatively connected to the cloud management module and is used to generate a unique corresponding text watermark through hash operation based on the identity information, instruction type and security requirements of the intelligent connected vehicle. The watermark embedding module is connected to the watermark generation module and the vehicle data interaction module respectively, and is used to embed the generated text watermark into the text data of the intelligent connected vehicle and output watermarked text data. The vehicle data interaction module is connected to the watermark embedding module, the watermark extraction module, the vehicle sensor, and the vehicle control unit, and is used to transmit the watermarked text data and receive external feedback data. The watermark extraction module is connected to the vehicle data interaction module and the security verification module, and is used to extract watermark information from the received watermarked text data. The security verification module is connected to the watermark extraction module, the cloud management module and the vehicle control unit, and is used to verify the extracted watermark information to determine whether the data has been tampered with, whether the vehicle identification is consistent with the vehicle identification stored in the cloud, and whether the timestamp is within a preset valid range. The cloud management module is connected to the watermark generation module and the security verification module, and is used to store the identity information, watermark template, and verification rules of the intelligent connected vehicle, and to synchronize watermark data and verification results.

2. The system according to claim 1, characterized in that, The watermark generation module uses a combination of encryption algorithm and hash operation to generate the text watermark, which includes a unique vehicle identifier, a data type identifier, a timestamp, and a verification code. The hash operation is represented as: (1) in, This is the watermark hash value. The unique identifier for intelligent connected vehicles. For timestamps, For data type identification, For an improved MD5 hash function, A unique key assigned to the cloud management module. The length of the watermark. This is an XOR operation.

3. The system according to claim 1, characterized in that, The watermark embedding module employs differentiated embedding strategies for different types of text data. For short instruction text, the text watermark is embedded in punctuation marks or spaces; for structured log text, the text watermark is embedded in redundant bits of the log field. The embedding process of the watermark embedding module is represented as follows: (2) in, For watermarked text data, The original text data, This is the original watermark information. For embedding strength, This is the watermark embedding function.

4. The system according to claim 1, characterized in that, The watermark extraction module uses an extraction algorithm corresponding to the embedding algorithm to locate the watermark embedding position according to the type of text data and extract the watermark information. The watermark extraction process of the module is described as follows: (3) in, For the extracted watermark information, This is the watermark extraction function. For watermarked text data, For embedding strength.

5. The system according to claim 1, characterized in that, The security verification module will extract the watermark information. The original watermark information stored in the cloud management module The comparison is performed to verify whether the vehicle identifier in the watermark is consistent with the vehicle identifier stored in the cloud and whether the timestamp is within the preset valid range. like If the vehicle identifier matches the vehicle identifier stored in the cloud and the timestamp is within a preset valid range, the verification passes; otherwise, the verification fails and a security warning is triggered. This is the preset allowable error.

6. A method for security protection of intelligent connected vehicles based on text watermarking, characterized in that, The method includes: A unique key is assigned to the intelligent connected vehicle through the cloud, the vehicle's unique identifier, current timestamp, and text data type identifier to be processed are obtained, a unique text watermark is generated through hash operation, and the text watermark is stored in the cloud; The vehicle-mounted terminal acquires the original text data, dynamically adjusts the embedding strength according to the text data type, and embeds the text watermark into the original text data to generate watermarked text data. The watermarked text data is transmitted to the target node, and the data is compressed and encrypted during the transmission process; The target node receives the watermarked text data and extracts the watermark information according to the text data type. The original watermark information is obtained from the cloud. The extracted watermark information is compared with the original watermark information. At the same time, it is verified whether the vehicle identification is consistent with the vehicle identification stored in the cloud and whether the timestamp is within the preset valid range. The verification result is then output. If the verification passes, the target node executes relevant instructions or stores data; if the verification fails, a security warning is triggered, abnormal data transmission is cut off, execution of relevant instructions is prohibited, and abnormal information is uploaded to the cloud.

7. The method according to claim 6, characterized in that, The hash operation is represented as: (1) in, This is the watermark hash value. The unique identifier for intelligent connected vehicles. For timestamps, For data type identification, For an improved MD5 hash function, A unique key assigned to the cloud management module. The length of the watermark. This is an XOR operation.

8. The method according to claim 6, characterized in that, The watermark embedding process is represented as follows: (2) in, For watermarked text data, The original text data, This is the original watermark information. For embedding strength, This is the watermark embedding function.

9. The method according to claim 6, characterized in that, The watermark extraction process is represented as follows: (3) in, For the extracted watermark information, This is the watermark extraction function. For watermarked text data, For embedding strength.

10. The method according to claim 7, characterized in that, The security verification rule is as follows: if the extracted watermark information With the original watermark information satisfy If the vehicle identifier matches the vehicle identifier stored in the cloud and the timestamp is within a preset valid range, the verification passes; otherwise, the verification fails. This is the preset allowable error.

Images